The development of vehicle controllers for autonomous racing is challenging because racing cars operate at their physical driving limit. Prompted by the demand for improved performance, autonomous racing research has seen the proliferation of machine learning-based controllers. While these approaches show competitive performance, their practical applicability is often limited. Residual policy learning promises to mitigate this drawback by combining classical controllers with learned residual controllers. The critical advantage of residual controllers is their high adaptability parallel to the classical controller's stable behavior. We propose a residual vehicle controller for autonomous racing cars that learns to amend a classical controller for the path-following of racing lines. In an extensive study, performance gains of our approach are evaluated for a simulated car of the F1TENTH autonomous racing series. The evaluation for twelve replicated real-world racetracks shows that the residual controller reduces lap times by an average of 4.55 % compared to a classical controller and even enables lap time gains on unknown racetracks.
IteraOptiRacing: A Unified Planning-Control Framework for Real-time Autonomous Racing for Iterative Optimal Performance
Shanghai Jiao Tong University(上海交通大学);University of Pennsylvania(宾夕法尼亚大学);University of Minnesota-Twin Cities(明尼苏达大学-双城分校);University of California, Berkeley(加州大学伯克利分校)
Two-timescale Mechanism-and-Data-Driven Control for Aggressive Driving of Autonomous Cars
Tsinghua University(清华大学)
由于存在显著的轮胎打滑,自动驾驶汽车的侵略性驾驶控制具有挑战性。用于在激进驾驶条件下对自动驾驶汽车进行建模和控制的数据驱动和基于机制的方法分别在数据效率和适应性方面受到限制。本文是对这两类方法融合的一种尝试。该方法采用基于机构和数据驱动的模块化设计,并考虑到汽车模型中的双时标现象,有效地提高了数据效率、传输能力和最终性能。在TORCS(The Open Racing Car Simulator)上验证了混合机制和数据驱动的方法。实验结果表明,我们的方法比纯粹的基于机制和纯粹的数据驱动的方法的好处。
RaceMOP: Mapless Online Path Planning for Multi-Agent Autonomous Racing using Residual Policy Learning
TUM School of Engineering and Design, Technical University of Munich(慕尼黑工业大学工程与设计学院);Graduate School of Information Science and Technology, The University of Tokyo(东京大学信息科学与技术研究生院)
Robust Modeling and Controls for Racing on the Edge
University of Pittsburgh(匹兹堡大学);University of Waterloo(滑铁卢大学);Massachusetts Institute of Technology(麻省理工学院);Carnegie Mellon University(卡内基梅隆大学);Argo AI
Model-Structured Neural Networks to Control the Steering Dynamics of Autonomous Race Cars
Professorship of Autonomous Vehicle Systems, Technical University of Munich(自动驾驶车辆系统教授职位,慕尼黑技术大学);Department of Information Technology and Electrical Engineering (DIETI), University of Naples Federico II(信息科技与电气工程系(DIETI),那不勒斯费德里科二世大学);Chair of Automatic Control, Department of Mechanical Engineering, TUM School of Engineering and Design(自动控制教授职位,机械工程系,慕尼黑技术大学工程与设计学院);Department of Industrial Engineering, University of Trento(工业工程系,特伦托大学)
The Autonomous Racing Software Stack of the KIT19d
Karlsruhe Institute of Technology(卡尔斯鲁厄理工学院)
Formula Student Driverless挑战工程专业的学生开发自动驾驶单座赛车,以寻求带来更多的毕业生,他们已经准备好解决与自动驾驶相关的现实世界问题。在本文中,我们介绍了KA-RaceIng参加2019年比赛的软件堆栈。我们涵盖了系统的基本模块,包括感知,定位,映射,运动规划和控制。此外,开发方法进行了概述,并给出了系统架构的概述。最后,我们提出了选定的运行时测量,数据日志,和竞争的结果,以提供一个深入了解最终原型的性能。
End-to-End Velocity Estimation For Autonomous Racing
Autonomous Systems Lab, ETH Zürich(苏黎世联邦理工学院自主系统实验室);Robotics and Autonomous Systems Group, CSIRO(澳大利亚联邦科学与工业研究组织机器人与自主系统组);Sevensense Robotics A.G(Sevensense Robotics 股份公司)
速度估计在无人驾驶汽车中起着核心作用,但标准和负担得起的方法难以应对极端情况,如由于存在高侧滑而导致的激进机动。为了解决这个问题,自动驾驶赛车通常配备昂贵的外部速度传感器。在本文中,我们提出了一个端到端的递归神经网络,该网络将可用的原始传感器作为输入(IMU,车轮里程计和电机电流)并输出速度估计。两个国家的最先进的卡尔曼滤波器,分别包括和排除昂贵的速度传感器的结果进行比较。所有方法都在一辆方程式学生无人驾驶赛车上进行了广泛的测试,该赛车具有非常高的侧滑(后轴10°)和滑移率(~20%),接近操纵极限。建议的网络是能够估计横向速度高达15倍,优于卡尔曼滤波器与等效的传感器输入和匹配(0.06 m/s RMSE)的卡尔曼滤波器与昂贵的速度传感器设置。
Autonomous Vehicles on the Edge: A Survey on Autonomous Vehicle Racing
University of Pennsylvania(宾夕法尼亚大学);ETH Zurich(苏黎世联邦理工学院);California Institute of Technology(加州理工学院);Technical University of Munich(慕尼黑工业大学);University of Virginia(弗吉尼亚大学);Clemson University(克莱姆森大学)
Steering Prediction via a Multi-Sensor System for Autonomous Racing
University of Burgundy(勃艮第大学);Sensors Group, Institute of Neuroinformatics, UZH/ETH Zürich(苏黎世大学/苏黎世联邦理工学院神经信息学研究所传感器组);University of Wurzburg(维尔茨堡大学);Université Rouen Normandie(鲁昂诺曼底大学);INSA Rouen Normandie(鲁昂诺曼底国立应用科学学院);Université Le Havre Normandie(勒阿弗尔诺曼底大学);Normandie Université(诺曼底大学联合体);LITIS UR 4108(LITIS UR 4108实验室)